my topic is about mitochondria and I'm keen on using bayesian data analysis but i don't know how to calculate sample size for this method
I personally would use standard procedures to estimate a sample size that would likely allow to reject H0 given a likely effect size and use this (particularily because the ethics commetee would expect this). The data may afterwards be analyzed using the Bayesian framework.
However, if you want to plan the study size right away based on Bayesian considerations, you have to write down how you want the posterior to look like, given a likely effect. You may then find - e.g. by simulation - the sample size that will lead to such a posterior (given a prior and sensible assumptions about the distribution of the values under the hypothesized effect). In simple cases and using conjugated priors/posteriors, you may even find an analytical solution.
I would suspect that it depends on what you want to do with the mitochondria. I would not choose a data analysis approach until I at least could specify a research question. Best, David Booth
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size